The overall hypothesis for this proposal is that in patients with ulcerative colitis (UC) the colonic epithelium expresses autoantigen(s) which ma drive the immune destruction of the epithelial cells. We earlier identified a putative colonic autoantigen, p40, that belongs to cytoskeletal tropomyosin (TM) family, and majority of UC sera react with p40/TM. Utilizing recombinant hTM isoforms (hTM-5) and isoform specific monoclinal antibodies (mAb) developed by us, we have observed that the normal colon epithelial cells (CEC) contain mostly hTM5 and hTM4 and not hTM1-3. We demonstrated autoantibody response in UC against hTM5 and hTM1. We will examine if there is any qualitative and/or quantitative differences in hTM isoform expression I CEC from UC compared. Using the hTM5 isoform, we identified two HLA DR2-binding peptides. We will focus on T cell responses against hTM5 and the peptides. We will also examine I cells responses against hTM4 and other hTM isoforms. For T cell responses, we will use both peripheral blood mononuclear cells (PBMC) and lamina propria lymphocytes (LPL). We have shown both in UC mucosa as well as in selected colon cancer cells that UC associated autoantibodies bind at the same site as the 7E12H12 mAb (developed by us earlier using highly enriched p40) does. This mAb has unique organ distribution such as colon, biliary tract, skin, eyes and joints. Recently we identified that 7E12H12 mAb reacts with unioval colonic epithelial protein (CEP) of Mr>200K. Preliminary experiments suggest that CEP may act as a "shuttle" to externus hTM5. We will critically examine the relationship between hTMs and CEP. Using purified CEP, we will also be cloned. Human colon cDNA expression libraries from normal and UC colon mucosa and cDNA libraries from two colon cancer cell lines are developed and will be utilized.